Initial push from clean repo

.gitattributes

@@ -57,3 +57,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+*.txt filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,41 @@
+---
+dataset_info:
+  features:
+    - name: text
+      dtype: string
+language_bcp47:
+  - zh-Hant-HK
+  - yue-Hant-HK
+license: cc-by-4.0
+pretty_name: Hong Kong content-based (Cantonese & Traditional Chinese) text corpus
+---
+
+# Dataset Card for `hk_content_corpus`
+
+## Dataset Description
+
+This dataset contains eight cleaned source-specific corpora of Hong Kong Cantonese and Traditional Chinese text, crawled from public websites and platforms.  
+Each file stores plain UTF-8 text, where **each record occupies one line**, and **blank lines serve as separators**.  
+This dataset is also available at Zenodo: https://doi.org/10.5281/zenodo.16882352  
+We only change file extension from .corpus to .txt here for HuggingFace's dataset viewer function  
+
+### Files
+
+| Filename                                   | Description                        | SHA256 checksum                                                    |
+|--------------------------------------------|------------------------------------|--------------------------------------------------------------------|
+| `appledaily_article_dedup.txt`             | Apple Daily news articles          | `0b6ad22b7a73230fd0e44af904c0cff1773cc871417f6ad3af11a783564bca15` |
+| `hkcitizenmedia_article_dedup.txt`         | HK Citizen Media articles          | `373a5b369d5e402e58760861e2bff2e618e7ee7fd4494988a39be1156f9dba84` |
+| `hkcnews_article_dedup.txt`                | Hong Kong Citizen News             | `67f909cfaf7d7a67df1dc79448f24622ec525a31b73b2e3293fbaad147470e69` |
+| `inmedia_article_dedup.txt`                | InMedia.hk articles                | `135251f3e8ba7587018b7b2814b9ea5bbebf69c98adca73d3ea4c9f9b5571957` |
+| `lihkg_posts_dedup_demoji_128.txt`         | Lihkg forum posts (emoji removed)  | `8e5e6de9c219aeccdaf13e9162b00cce2eeb7f595023a7bf19d4b5660395a3ee` |
+| `openrice_review_dedup_demoji.txt`         | OpenRice user reviews              | `dd5835a7effe49bb96a31e0c0eab43dea17ef23b3e1d9cefdc186aca276897ce` |
+| `thestandnews_article_dedup.txt`           | Stand News articles                | `847ef0f5809481caf4bf21100e4b34207513d724dafb9f42b162ef49079e7dba` |
+| `wiki_hk.txt`                              | Wikipedia (zh-hk)                  | `bd33008802797b33df8484cf1113be6a0b38547fe13515f5c4edbf4ccad270db` |
+
+## Intended Uses
+
+- Language model pretraining and finetuning  
+- Hong Kong Cantonese/Traditional Chinese linguistic modeling  
+- Downstream tasks such as classification or generation
+
+NOTE: HKNSL became effective since 2020-6-30, which can create bias on user content created afterwards. Those portion of data should be used with caution.

appledaily_article_dedup.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b6ad22b7a73230fd0e44af904c0cff1773cc871417f6ad3af11a783564bca15
+size 3143931888

dataset.py

@@ -0,0 +1,50 @@
+import datasets
+
+_DESCRIPTION = "Cleaned Hong Kong Cantonese / Traditional Chinese corpora from multiple sources."
+_HOMEPAGE = "https://huggingface.co/datasets/SolarisCipher/hk_content_corpus_script"
+_CORPUS_FILES = {
+    "appledaily_article": "appledaily_article_dedup.txt",
+    "hkcitizenmedia_article": "hkcitizenmedia_article_dedup.txt",
+    "hkcnews_article": "hkcnews_article_dedup.txt",
+    "inmedia_article": "inmedia_article_dedup.txt",
+    "lihkg_posts": "lihkg_posts_dedup_demoji_128.txt",
+    "openrice_review": "openrice_review_dedup_demoji.txt",
+    "thestandnews_article": "thestandnews_article_dedup.txt",
+    "wiki_hk": "wiki_hk.txt",
+}
+
+class HKContentCorpus(datasets.GeneratorBasedBuilder):
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=datasets.Features({"text": datasets.Value("string")}),
+            supervised_keys=None,
+            homepage=_HOMEPAGE,
+            citation="",
+        )
+
+    def _split_generators(self, dl_manager):
+        return [
+            datasets.SplitGenerator(
+                name=split_name,
+                gen_kwargs={"filepath": filepath},
+            )
+            for split_name, filepath in _CORPUS_FILES.items()
+        ]
+
+    def _generate_examples(self, filepath):
+        with open(filepath, encoding="utf-8") as f:
+            buffer = []
+            example_id = 0
+            for line in f:
+                line = line.strip()
+                if line:
+                    buffer.append(line)
+                else:
+                    if buffer:
+                        yield example_id, {"text": "\n".join(buffer)}
+                        example_id += 1
+                        buffer = []
+            if buffer:
+                yield example_id, {"text": "\n".join(buffer)}

hkcitizenmedia_article_dedup.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:373a5b369d5e402e58760861e2bff2e618e7ee7fd4494988a39be1156f9dba84
+size 1957068

hkcnews_article_dedup.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67f909cfaf7d7a67df1dc79448f24622ec525a31b73b2e3293fbaad147470e69
+size 81680237

inmedia_article_dedup.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:135251f3e8ba7587018b7b2814b9ea5bbebf69c98adca73d3ea4c9f9b5571957
+size 340775109

lihkg_posts_dedup_demoji_128.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e5e6de9c219aeccdaf13e9162b00cce2eeb7f595023a7bf19d4b5660395a3ee
+size 3131803511

openrice_review_dedup_demoji.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd5835a7effe49bb96a31e0c0eab43dea17ef23b3e1d9cefdc186aca276897ce
+size 1572247635

thestandnews_article_dedup.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:847ef0f5809481caf4bf21100e4b34207513d724dafb9f42b162ef49079e7dba
+size 220252263

wiki_hk.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd33008802797b33df8484cf1113be6a0b38547fe13515f5c4edbf4ccad270db
+size 1653596553